Instructions to use Myric/granite-4.0-h-tiny-APEX-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Myric/granite-4.0-h-tiny-APEX-GGUF
# Run inference directly in the terminal:
llama cli -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Myric/granite-4.0-h-tiny-APEX-GGUF
# Run inference directly in the terminal:
llama cli -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Myric/granite-4.0-h-tiny-APEX-GGUF
# Run inference directly in the terminal:
./llama-cli -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Myric/granite-4.0-h-tiny-APEX-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Use Docker

docker model run hf.co/Myric/granite-4.0-h-tiny-APEX-GGUF

LM Studio
Jan

vLLM

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Myric/granite-4.0-h-tiny-APEX-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Myric/granite-4.0-h-tiny-APEX-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Myric/granite-4.0-h-tiny-APEX-GGUF

Ollama
How to use Myric/granite-4.0-h-tiny-APEX-GGUF with Ollama:
```
ollama run hf.co/Myric/granite-4.0-h-tiny-APEX-GGUF
```
Unsloth Desktop

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Myric/granite-4.0-h-tiny-APEX-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Myric/granite-4.0-h-tiny-APEX-GGUF with Docker Model Runner:
```
docker model run hf.co/Myric/granite-4.0-h-tiny-APEX-GGUF
```

Lemonade

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Myric/granite-4.0-h-tiny-APEX-GGUF

Run and chat with the model

lemonade run user.granite-4.0-h-tiny-APEX-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Myric/granite-4.0-h-tiny-APEX-GGUF

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Myric/granite-4.0-h-tiny-APEX-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Myric/granite-4.0-h-tiny-APEX-GGUF

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Myric/granite-4.0-h-tiny-APEX-GGUF" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Myric commited on Jul 23

Commit

44e9d5d

verified ·

1 Parent(s): 79ffe70

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +49 -0

README.md CHANGED Viewed

@@ -35,8 +35,15 @@ Perplexity on wikitext-2-raw (test, 200×512-token windows), `llama-perplexity`.
 |------|------|-----|-----|-----------|
 | bf16 (reference) | 13 GB | 16.0 | 8.868 | — |
 | **APEX i-quality** (ssm@Q6) | **4.4 GB** | 5.40 | **8.901** | **+0.38%** |
 | APEX hand-roll (ssm@Q8) | 4.5 GB | 5.54 | 8.913 | +0.51% |
 Within **0.38% of full-precision perplexity at ~3× smaller**, and it runs comfortably
 on modest hardware (~14 GB bf16 → 4.4 GB). Coherent, ~110 tok/s on a single GPU.
 Built with a diverse imatrix (Bartowski `calibration_datav3`), full expert coverage.
@@ -71,6 +78,48 @@ the linear-recurrence state helps: it doesn't — it's **larger and slightly wor
 hybrid architectures the SSM/recurrence tensors just aren't precision-sensitive. The
 hand-roll is included to document the experiment; `i-quality` is the recommended tier.
 ## Attribution & licenses
 See [`LICENSE`](LICENSE) (Apache-2.0) and [`NOTICE`](NOTICE).
 - Base: **IBM** ([@ibm-granite](https://huggingface.co/ibm-granite)) — [granite-4.0-h-tiny](https://huggingface.co/ibm-granite/granite-4.0-h-tiny) (Apache-2.0)

 |------|------|-----|-----|-----------|
 | bf16 (reference) | 13 GB | 16.0 | 8.868 | — |
 | **APEX i-quality** (ssm@Q6) | **4.4 GB** | 5.40 | **8.901** | **+0.38%** |
+| **APEX i-quality — torch-imatrix** | **4.4 GB** | 5.40 | **8.864** | **−0.04%** |
 | APEX hand-roll (ssm@Q8) | 4.5 GB | 5.54 | 8.913 | +0.51% |
+The **torch-imatrix** row is the *same* APEX i-quality recipe, quantized with an
+importance matrix computed independently in PyTorch (see [*The torch-imatrix
+variant*](#the-torch-imatrix-variant)) instead of `llama-imatrix`. Its PPL (8.864)
+is inside the ±0.11 error bar of both bf16 and the standard i-quality — i.e. the
+two imatrices are interchangeable in quality.
 Within **0.38% of full-precision perplexity at ~3× smaller**, and it runs comfortably
 on modest hardware (~14 GB bf16 → 4.4 GB). Coherent, ~110 tok/s on a single GPU.
 Built with a diverse imatrix (Bartowski `calibration_datav3`), full expert coverage.
 hybrid architectures the SSM/recurrence tensors just aren't precision-sensitive. The
 hand-roll is included to document the experiment; `i-quality` is the recommended tier.
+## The torch-imatrix variant
+`granite-4.0-h-tiny-APEX-i-quality-torch.gguf` (+ `granite-4.0-h-tiny-torch.imatrix`)
+is a companion build that answers one question: **can the calibration imatrix be
+produced without llama.cpp, and does that change the result?**
+**What it is.** An importance matrix (imatrix) records, per weight tensor, the
+per-input-channel sum of squared activations over calibration text — it tells
+`llama-quantize` where to spend bits. Normally you get it from `llama-imatrix`,
+which needs the whole model resident in RAM+VRAM. Here it was instead computed
+**directly from the Hugging Face model in PyTorch**, by registering forward hooks
+on every matmul (attention, Mamba-2 in/out projections, router, shared experts,
+and per-expert routed FFNs) and accumulating Σx² as calibration text streams
+through. The generator is **band-serialized**: it loads a few decoder layers at a
+time, runs all calibration chunks through them, caches activations, frees them,
+and moves on — so peak VRAM is a few layers, **not the whole model** (this run
+peaked at ~6.4 GB on a 16 GB card, vs the ~14 GB a full-resident forward needs).
+That decouples imatrix generation from llama.cpp's memory model: you can calibrate
+a model far larger than your GPU by streaming it in bands.
+**Everything else is identical.** Same bf16 base, same APEX `--tensor-type-file`
+recipe, same `calibration_datav3`, same 126×512 calibration windows. The *only*
+difference from the standard `i-quality` file is which tool produced the imatrix.
+The two quants therefore differ **only** in the tensors whose quantization
+consumes an imatrix (the K-/I-quant expert, attention and Mamba weights); the
+Q8_0 shared experts and F32 tensors are bit-identical.
+**Validation.** The PyTorch imatrix matches `llama-imatrix`'s output tensor-for-
+tensor (368/368 names) with **median per-tensor correlation 0.995**. The handful
+of lower-correlation tensors are the post-nonlinearity inputs (`ffn_down_shexp`,
+`attn_output`) where transformers' naive Mamba scan and SiLU-gated intermediates
+differ numerically from llama.cpp's kernels — a known wrinkle that does not affect
+quality, since the quantizer only needs *relative* per-channel importance. The
+proof is in the table above: the torch-imatrix quant scores **PPL 8.864**, inside
+the ±0.11 error bar of both bf16 and the standard i-quality. The two imatrices are
+interchangeable.
+**Pick whichever you like** — they are equivalent in quality. The standard
+`i-quality` is the reference; the `torch` build is provided for anyone who wants a
+llama.cpp-free, VRAM-bounded path to the same result (e.g. calibrating very large
+MoEs on modest hardware).
 ## Attribution & licenses
 See [`LICENSE`](LICENSE) (Apache-2.0) and [`NOTICE`](NOTICE).
 - Base: **IBM** ([@ibm-granite](https://huggingface.co/ibm-granite)) — [granite-4.0-h-tiny](https://huggingface.co/ibm-granite/granite-4.0-h-tiny) (Apache-2.0)